Utterance comprehension in spontaneous speech
Phonetic reductions and lexico-grammatical context
The nature of spontaneous speech raises questions concerning the relationship between phonetic
reductions and unimpaired comprehension of an utterance. The results of a pilot experiment based on individual
production and community perception in Czech spontaneous conversations suggest that there is detectable patterning in
how and where speakers tend to save articulatory effort without compromising listeners’ comprehension. We identify
several non-phonetic factors that appear to motivate the reductions and their position in utterances: information
structure, semantic content, morphosyntactic redundancy, and syntactic organization. Based on our findings, we suggest
that constructional representations should incorporate the ways in which grammatical organization and phonetics may
‘fill in’ for each other in the course of the collaborative interpretive effort during spontaneous interaction.
Article outline
- 1.Introduction
- 2.Data collection and methodology
- 2.1Word-reduction-rate method
- 2.2Experimental probe: The relationship between reductions and comprehensibility
- 2.2.1Step one: Distribution of reductions across utterances
- 2.2.2Step two: Word identifiability out of context
- 3.Results: WRR distribution trends
- 3.1Reduction rates in production
- 3.2Word identifiability vis-à-vis utterance comprehension
- 3.3Word identifiability vis-à-vis respondent success rate
- 4.Potential motivations for phonetic reductions
- 4.1Lexical category
- 4.2Discourse-based motivation
- 4.2.1Information structure and WRR trajectories
- 4.2.2Utterance boundaries
- 4.3Reductions vis-à-vis morphology
- 4.4Phonetic chunks and their lexico-syntactic conditioning
- 4.4.1Candidates for phonetic chunks
- 4.4.2Relationship between WRR and syntactic clause boundaries
- 5.Challenges for a constructional account of spontaneous speech
- 6.Summary and conclusions
-
Notes
-
Glosses used throughout the chapter
-
References
-
Appendix